A Hypertext Approach to Foreign Language Reading: Student Attitudes and Perceptions

Jeong-Bae Son
The University of Southern Queensland

ABSTRACT

This article reports the results of a study conducted to examine the use of three different reading text formats: paper-based format (PF), computer-based non-hypertext format (NHF), and computer-based hypertext format (HF). It investigates foreign language learners' reactions to the three text formats, focusing particularly on the usefulness of hyperlinks in computer-mediated text to provide readers with optional assistance during independent reading. Data collected from an interview with a group of Korean as a foreign language (KFL) students are presented and discussed. The results of this study show that the students considered the use of hyperlinks to be helpful and useful for their learning and suggest that it is important to look at the way in which reading materials and supporting information are presented when designing or selecting computer-assisted language learning (CALL) programs.

INTRODUCTION

Of all the areas in which computers have been used to teach language skills, reading is one of the language activities which has received great attention in research and instruction. One of the approaches which is being discussed for effective computer-assisted reading practice and is becoming significant in recent computer-assisted language learning (CALL) activities is the use of hypertext (Ashworth, 1996; Biber, 1992; Donaldson, Haggstrom & Morgan, 1994; Evans, 1993; Fox, 1990; Hult, Kalaja, Lassila & Lehtisalo, 1990; McVicker, 1992; Sengupta, 1996; Son, 1998). This article presents the results of an interview study that examined the different aspects of three presentation methods of reading passages and lexical resources in three different text formats: a paper-based format (PF), a computer-based non-hypertext format (NHF), and a computer-based hypertext format (HF). With a special focus on the use of hyperlinks (or hypertext links) for presenting lexical information, the study specifically explored on-line lexical resources of hypertext-based courseware for learning Korean as a foreign language (KFL) and investigated students' attitudes toward the three reading text formats and perceptions of reading linear texts and non-linear texts.

HYPERTEXT AND CALL

Hypertext can be simply defined as "non-sequential text which is organised to allow readers/users access to non-linear information" (Son, 1998, p. 115). Similarly, Slatin (1991, p. 56) defines hypertext as follows: "A hypertext (or hyperdocument) is an assemblage of texts, images, and sounds - nodes - connected by electronic links so as to form a system whose existence is contingent upon the computer". Tolhurst (1995, p. 22) also argues that "hypertext can be viewed functionally as nodes of information that are linked, allowing readers to follow a variable reading path of associations based on semantic links". On the other hand, Whalley (1993, p. 8) sees hypertext critically as "a fragmented text form whose components can be rapidly accessed". He (1993, p. 11) points out that hypertext works for some aspects of learning and teaching, especially where text is a "database of facts", but not where text is a "configuration of ideas" presenting a writer's cohesive argument. With a particular focus on the advantages of connectivity of hypertext, Landow (1990, p. 42) claims that hypertext promotes collaborative learning in a situation where it is not important that "readers retain less information they encounter while reading text on a screen than that they read on the printed page".

A number of CALL studies (e.g., Armstrong & Yetter-Vassot, 1994; Ashworth, 1996; Biber, 1992; Hult et al., 1990; McVicker, 1992) suggest that the use of hypertext has great potential in language learning, and hypermedia applications can enrich learning tasks by providing learners with various types of on-line information. In this regard, Evans (1993, pp. 214-215) notes, "The integration of text, sound and visual data clearly is of great benefit to the learner as this will reinforce comprehension, pronunciation and contextual use in a way that traditional materials are not able to do". The most common activity in a hypertext application consists of "moving the cursor to a word on the screen, pressing a given key and getting the available information" (Leffa, 1992, p. 66). The linked pieces of information can be presented in a pop-up window or in a separate page on the same screen or a different screen. Hypertext systems can make an explanation of any word pop up in a separate window when learners click on any word that they find unfamiliar while allowing learners themselves to decide how much help they need and where they can finish their exploration in relation to the specific word or text (Son, 1998).

While many researchers (e.g., Chun & Plass, 1996a, 1996b; Hulstijn, 1993; Knight, 1994; Leffa, 1992; Laufer & Hill, 2000; Lomicka, 1998; Lyman-Hager, Davis, Burnett & Chennault, 1993) conducted studies on the effects of electronic dictionary or glossary use or look-up behaviour of readers on vocabulary learning or reading comprehension without any special focus on the use of hyperlinks in the presentation of dictionaries or glossaries, some other researchers

In a comparative study of hyper-reference and conventional paper dictionary use on the measures of consultation frequency, study time, efficiency and comprehension, on the other hand, Aust et al. (1993) found that their subjects were generally enthusiastic about the use of hyper-reference dictionaries. Hyper-reference users could consult definitions for any word in an electronic book, and after clicking on a word, the definition window immediately appeared on the page opposite to the selected word. The results indicated that users of hyper-references consulted over two times as many definitions as users of conventional dictionaries, and users of bilingual dictionaries consulted 25% more definitions and completed reading in 20% less time than did users of monolingual dictionaries. However, there were no significant differences between hyper-reference and paper media and between bilingual and monolingual dictionary use in comprehension measured by a proposition recall protocol. De Ridder (2000) also examined the effects of highlights indicating a link toward dictionary definitions on the reading process. Seventeen Dutch learners of French individually read two glossed texts in French under two conditions: one in which the glossed words were highlighted and another one in which the words were not marked in any specific way. From the results of vocabulary and comprehension tests and the total amount of time spent on clicking and reading, De Ridder found that reading a text in a marked condition is fundamentally different from reading the same text in a condition without highlights. The students felt conditioned to follow the links presented in the marked text, which resulted in significantly higher scores on the receptive vocabulary test taken immediately after reading and more clicking time in the marked condition than in the unmarked condition.

METHOD

Design

The study reported in this article is part of a broader-based study on the relative effectiveness of the use of three different reading text formats (i.e., PF, NHF & HF) in the foreign language classroom, which included the quantitative analysis of vocabulary and reading tests and the qualitative analysis of questionnaires and an interview. The focus of this article is a semi-structured interview with KFL students who enrolled in a second year Korean language course at an Australian university. The 14-week course consisted of 8 contact hours per week, comprising a one-hour lecture and seven tutorial hours. A one-hour reading session was one of the seven tutorial hours that students were required to attend each week. The study was conducted during the reading sessions in a computer lab where there were 15 Macintosh computers. CALL materials for the study were accessed and used in the lab during the reading sessions only.

The students were divided randomly into three groups, each using different text formats. The content of the text materials used by each group was identical. That is, the students read the same reading passages in three reading conditions differentiated by three text formats. Each group was allocated one of the three different text formats for three sessions over a period of three weeks. After each three-session period, there were revision sessions covering the preceding three sessions. In total, there were three experiments during the study, each three-session period being designated as one experiment. The formats were rotated around the groups so that, through the three experiments, each group was able to experience all of the formats. The sizes of the student sample were determined by the actual student enrolments in the available classes.

In terms of CALL applications, the hypertext concept could be implemented by the format of stand-alone courseware or Web documents on the Internet. In order to focus on the use of hyperlinks in reading texts rather than on the use of the Web technology which involves issues of connection and retrieval time on the Internet, the study reported in this article used courseware installed in the hard disk of a server computer. The design of the hypertext-based courseware comprised an electronic glossary that employed hyperlinks to present vocabulary and grammar information.

Subjects

All of the subjects in this study expressed their willingness to participate in the study and filled out a consent form and an information questionnaire on their personal background. Experiment 1 was conducted with 15 students (9 females and 6 males; mean age 20, ranging from 18 to 26 years) over three consecutive weeks during three class periods. Of the 15 students, 12 students' L1 was English

Materials

A total of nine reading passages were prepared before the study. They were specially written by the author for both the course and the study, and then checked by two other Korean teachers who were also involved in the Korean course to ensure the appropriateness of the difficulty level for the students and the correctness of the reading passages. In the development of the passages, it was also necessary to integrate the reading practice, using these passages, into the existing curriculum, so most topics and patterns of the passages were selected from the students' main textbook, Learning Korean: New Directions - Advanced Study Modules 1, 2 & 3 (Buzo & Shin, 1994), which was used for the Korean course. Each reading passage contained approximately eight Korean sentences consisting of 90 Korean words on average and was used for one unit during one session.

A high-frequency vocabulary list compiled by the author was distributed to the students in the first session of the course. The list was a collection of the most common words to be found in several Korean textbooks used in the first year Korean programs in Australian universities. The students were allowed to refer to the list during reading sessions because the vocabulary information accompanied by the reading passages for the study provided only uncommon words or new words which did not appear on the list. This free access to the list served to allow the students to prepare themselves before and during the reading sessions by giving them time to review words that they had learnt before. To understand the reading passages, the students had to rely heavily on the glossaries presented on the target text format.

A total of nine worksheets in PF were used for nine sessions. Each worksheet contained a Korean reading passage and a list of vocabulary and grammatical patterns. The passage was typed on the upper half of the page. The glossary section on the lower half of the page provided context-specific definitions for all of the new words and patterns. This structure gave the same direct access to the glossary in the off-line reading as in the on-line reading without using paper dictionaries.

In RE Ga, which was structured in a non-hypertext format, glossaries and grammar points were provided on separate screens when the students clicked on a certain icon on the main text screens. RE Na, on the other hand, employed the concept of hypertext so the reading passages and glossaries were linked in a non-linear way. The students were able to get the meanings of new words instantaneously and directly from the text pages while moving the mouse on the screen. Unlike the paper-based format and the non-hypertext format, there were also options for choosing more information on selected grammar points with the mouse as an indicator. From a gloss window, the students could access an information window via the 'More' button activated by a mouse click. The information window displayed information on the selected pattern with a grammatical explanation and an example sentence to give an idea of the usage of the pattern. By pressing the 'O.K.' button, the students were led to the main text screen. Whatever option they chose, it appeared on the same screen while displaying both the text being read and the glossary information at the same time.

After using the three different formats of reading texts, the students were asked to take follow-up tests in each session as part of classroom activities. The tests were achievement tests and included, first, a vocabulary test that measured the students' vocabulary recognition and, second, two types of reading comprehension tests (i.e., a multiple-choice comprehension test to measure their understanding of the text and a recall test to measure their identification of idea units in the text.). At the end of the course, a post-questionnaire was given to the students to document the students' attitudes toward computer-assisted reading activities as well as the students' evaluation of the computer programs they used during the course.

In addition, there was an interview which lasted about 15-20 minutes per student

1. Which text format do you think was the most helpful to you in understanding the reading passages? Why?
2. What do you think about the method of listening to the words and sentences in the computer programs whilst working with the reading passages? Did it help you?
3. Do you think you used different strategies to approach the reading passages of the three reading text formats? Can you describe the strategies you used to work with each text format?
4. Do you have any other comments to make about the reading passages, your preferences in the reading text formats, and your test results?

Question 4 was asked after showing and explaining to the students the results of their vocabulary, reading comprehension, recall and cloze tests. Although the students were asked four main questions in the interview, the replies were not restricted in length and topic. The interview allowed the students to speak freely about their perceptions and experiences relating to the research questions.

RESULTS

This section presents the data collected from the interview which was recorded with each participant's permission and later transcribed. In the presentation of the data, the participants' actual words are conveyed as central so as not to ignore any particulars that might be related to the results of the study. The interview responses are categorised and presented here under the four main interview questions.

The most helpful format in understanding the reading passages

In terms of the most helpful format, most students chose HF (5 students), followed by PF (3 students) and NHF (1 student). The students who chose HF pointed out easy and instant access to the vocabulary information in HF as the main reason for their choice:

Student A1: In understanding them, I suppose this one - the hypertext. Because, well, I mean, with these two [PF and NHF] you have to go and actually look for the meanings. And by then, you know, I mean, there's time, and understanding. But this one [HF] the meaning came up straight away, you could just click and insert English meanings straight to the sentence and you have it right there.

The students who selected PF mentioned that they could make notes and see all the information at the one time:

Student A3: Because it's easy to use and easy to see the vocabulary.

The student who chose NHF expressed his general impression of NHF instead of particular reasons for his choice:

Student C1: Because it gave you a general idea of what the whole paragraph, the whole text was about. ... As you work through, any word you are unsure of, you could find in the dictionary or whatever it was.

The usefulness of the hypertext was also observed in his words:

Student C1: This hypertext is good because you could just read, you could just read straight through and any words that you wanted to associate with sounds you could just click on the sounds, click on the words so it's good for association with words and things like that.

In addition, he complained that there was no support for the sounds in PF:

Student C1: The paper-based, just worked through from start to finish, just associated words but you couldn't use the sounds, you couldn't use the sounds to help and all sorts of things.

Student A1: The thing that annoyed me about this one [NHF] is that you have to change screens so you couldn't, because you couldn't write on the computer screen, you couldn't write the meaning above words, and so you have to go to the other screen and you have to memorize the word and you have to click back again. So that was annoying.

However, there were some students who thought that NHF was good in terms of memorizing the words presented in the vocabulary section:

Student A1: It [NHF] was good because I had remembered the words whereas this one [HF] when the words just came up automatically I just didn't look at the Korean words any more. I just learnt the English words which I already knew. So I didn't learn any vocab from this one [HF].

In brief, the students' responses indicate that HF was the most helpful format in understanding the reading passages. The students felt that HF, which provided immediate access to adjunct information, helped their reading. While the students valued making notes in PF, they generally disliked changing screens in NHF although some of them thought that NHF was beneficial for vocabulary learning in terms of making them to memorize words.

All of the students expressed the view that listening to the words and sentences in the computer programs was helpful:

Student A2: Yeah, very, I mean, it's very helpful. Because you get to know how to request a favour then and how they practice sounds when you look at them. Sometimes it's very hard because they don't sound the same as they look. So, that was easy.

However, two students noted that they did not use the listening features when they worked with the computer programs because of a personal decision and a working style:

Student A1: Oh, I didn't use it at all the whole time. Uh, it kind of wasn't the objective of the exercises the way we did. I didn't really come into it. It would be helpful. I can imagine it would be very helpful. But I didn't use it.

In short, the students' responses indicate that the availability of the listening function appears to be valued by the students. Most students made use of the listening option provided in the computer programs.

Of the nine students, eight students responded that they used different strategies for each text format. However, it seemed that the students did not identify exactly what kinds of strategies they adopted in working with each text format. Instead, they tended to simply describe their techniques for using lexical resources in the three different formats. With PF, most students tried to write the meanings of the words and sentences in the passages in English or make their own notes on the printed pages:

Student A1: I kind of was not in very good practice so I was writing English words about Korean words when I didn't know the word. That was really the only way I could have approached to that one. And each text, all of them I wrote the English translation, I didn't just read through it because I can't remember anything that way and I don't learn.

With NHF, most students were more concerned with the learning of vocabulary than the understanding of the passages so they tried to use the on-line help on new words and patterns. However, they felt that it was annoying and time consuming because they had to go back and forth between the passage screen and the vocabulary list screen:

Student A1: This one, the non-hypertext one, ... Usually, if there were couple of words I didn't know I'd try to remember both of them from the dictionary screen and then click back again but it was really annoying actually having to click back again between the two.

Because of the need to change screens, one student even confessed that she did not want to use the on-line dictionary while reading the passages:

Student B1: With the non-hypertext, if I didn't need to use the dictionary, I didn't because it was a bit annoying.

The students expressed the view that it was easy to use HF and they could read the passages quickly with the aid of glossaries which appeared instantly on the same screen. They also made positive comments regarding the options for choosing more information on particular grammatical points:

Student B2: With the hypertext one, I think that was best because you go at your own pace not only just get the vocab but also get sentence structures and that, which helped understand the sentences and how to say and use practical sentences or you just get grammatical patterns and you have to remember and understand how to use in practical sentences.

However, two students pointed out that they used HF in a lazy way by simply looking at the English translations rather than trying to work hard at memorizing the meanings of the Korean words or sentences:

Student A1: The hypertext format, I must admit I approached with a really lazy manner because I knew I just put the mouse on the words, underline and just went along. And if there was a translation I'd look at it and if there wasn't I'd use the word anyway and I just wrote that one. I was really lazy on that one. And, I could go through these ones very quickly.

One student said that she used similar strategies with all text formats. The student tried to translate the passages with the glossaries provided and to write down everything in English. Interestingly, the use of different paths to the vocabulary information was not a reason for the use of different strategies of working with each text format to this student:

Student C2: With the paper one, I actually sat there and I just went through. With nearly all of the types of formats I went through and I actually tried to translate it into English and I'd sit and write down everything so that I could understand it more. For this one [NHF] I sat and went through and where I didn't know a word I'd look it up. And it was the same for the hypertext ones except that I'd have to keep referring to the other page. And this one [HF], as well. So basically they were all the same. It's just that I was referring to different parts to find words I didn't know. Yeah, very similar.

On the whole, most students tended to use different strategies to approach the reading passages in the three different formats, mainly because of the different presentation methods that were embedded in each text format. It seemed that HF facilitated the reading process of the students by enhancing strategies in a self-directed way although it also allowed some students to work in an effortless manner.

With regard to the relationship between the preference and performance of students, the interviewees gave thinkable reasons for their test results with general comments on how they felt about the text formats or the reading practice. For example, Student A1 explained why her overall scores on the tests improved as she went along:

Student A1: Well, I think I can see why I got all my results on the ones I did. I say my paper ones were probably pretty poor in some areas, especially the recall test, because at that stage I hadn't worked out my techniques counting the number of words and sentences and memorizing the texts after it. That's how I can see I got the very lowest score first. And I think, because I sort of improved as I went along overall, I think. I think that's the reason. Just getting used to it week by week.

In addition to the above comment on her performance, she also indicated alternative directions for further experiments:

Student A1: So I think it would be interesting to see the other way around if I hadn't started with the paper. And the other thing that is difficult, also is, because they are so much standard, I mean, I got mine, the order of these, all chopped and changed, they didn't just progress and get progressively harder. So that would have a lot to do with scores, I think also.

Student B2 compared the use of PF and NHF in relation to his scores on the tests:

Student B2: I think, with that one, with the paper-based format, I think, maybe because, they are, I actually had a good look at the vocab, that week, and maybe that helped out as well. And I spent a lot more time with the non-hypertext one. I just tried to understand the whole sentence structures rather than just understand, just remember the vocab and that.

Student C3 made positive comments on the use of HF:

Student C3: I can understand why I got a better mark for the hypertext than I would have for anything else. Umm, because, umm, yeah, I could follow the sentences as it ran on the left. Because I write it down as well. It more, it probably just seemed to sink in more than other words whereas otherwise for the paper one I was just writing down sentences about it, looking more at the Korean words, the sentences and program bit.

For Student B3, NHF provided the greatest difficulty in terms of focusing on the reading texts. He pointed out disadvantages in the use of NHF in relation to his low scores on the tests achieved from working with NHF:

It seemed to be natural to Student C2 that she got better scores with PF:

Student C2: I think, the one that I preferred doing and the one that I could just sit down and doing was the paper-based format, and that is sort of showing in the marks.

In her remarks, there was an interesting indication that she enjoyed the process of the experiments as well:

Student C2: Doing it on the computer screens using reading passages, having vocab lessons and stuff like that, it's good idea and it's very helpful because it helps you learn quickly. And I found also the way that you taught the class as well, you'd have given some time to read things and read passages and then write down a list of the things. That sort of made us have to sit down and learn the vocab before we did the test which was good because we are not marked on it. But it was helping us just sort of trying a little bit harder on the vocab. It was a really good idea.

As implied by the above words, the students tended to agree with the assumption that their test results are closely related to their preferences in the reading text formats, even though there were some unexpected results:

Student A1: Umm, I am surprised that I got the highest result for the cloze test, you know, in the last one. Because that's one, I thought, that I wasn't reading properly. I am really surprised at that. I don't know, maybe it's just by the end of semester, I mean, probably you're getting better at reading Korean and getting better at reading the texts, truly. I don't know. I am really surprised.

Student B3: The only thing, I think the score, high score reflects the ability to remember in English which isn't necessarily good in language. In the long run, I might remember more what Korean content means than I do in the hypertext because I am using my English brain, not my Korean brain.

On the question of whether the difficulty of the reading passages influenced their test results, five students did not think it affected their performance:

Student B1: I don't think so. Because with some of the questions you really didn't have to know exactly what it meant. Because like, I think, the recall test, you just have to write in English, yeah.

Although the difficulty levels of the reading passages were carefully controlled in terms of the number of sentences and the complexity of sentence structures, the rest of the students (four students) expressed the view that there were some easy and difficult passages and the difficulty of the reading passages might affect their results:

Student C2: Because there were some chapters in the book that have a lot of vocab and a lot of like very difficult to learn vocab. Umm, so I think that, sometimes, you know, that could affect it as well.

In summary, most students agreed that the test results reflected their performance and perception in each text format. The students' responses indicate that the reason they gained the lowest scores on the tests when they worked with NHF was that they did not like the way they had to work with it.

DISCUSSION

The study was conducted in a foreign language classroom equipped with computers. University students learning KFL were given identical Korean passages to read in three text formats. In PF, the students read the target passages on printed pages accompanied by glossaries. In NHF and HF, they read the passages on computer screens that offered on-line glossaries with sound.

The results of the study indicate that, due to the ease of working on the target text with a reference, the students were able to focus on the target text more in HF and PF than in NHF. There were similar degrees of motivation in working with HF and PF. The students seemed less motivated by NHF because they did not like getting adjunct information on the target text by changing screens, which they found annoying and time-consuming.

PF was the format that the students reported they felt was easier to use because they were most familiar with it and the format allowed the students to make notes on the text. While the students regarded the information in HF as helpful, most students pointed out that they needed to write the meaning of the text on paper in order to study the text. Presumably, the students might need a certain period of time to get used to the instructional treatment of HF before they can take advantage of the instructional approach HF offered. These findings suggest that using HF and PF together might be the best way of helping students read and understand the text, provided that there is no time pressure.

In relation to the question of what students think about hyperlinks, this study found that the students considered the use of hyperlinks as helpful and useful for their learning. It also showed that the students had a preference for HF in comparison with NHF. The students tended to actively use the hyperlinks embodied in the HF program and demonstrated positive attitudes toward hypertext applications and computer-assisted reading activities.

These findings suggest that it would be worthwhile to use hyperlinks in the presentation of on-line references. The findings also support the idea that hypertext can assist language learning because of its linking abilities. Hypertext is likely to facilitate a discovery approach, enrich learning experiences and enhance learning strategies in a self-directed way. Certainly, no hypertext applications can be expected to be useful and helpful for all FL learners in all circumstances. Rather, the results of this study suggest the need for further investigation of hypertext applications. In order to find a way to enhance the design of FL courseware, this study was undertaken with specially designed courseware only. It would be valuable for future research to investigate the use of hyperlinks in other contexts such as on the World Wide Web.

Further investigations are surely needed to compare and evaluate paper-based formats and hypertext-based formats in various ways as well as to determine

CONCLUSIONS AND IMPLICATIONS

Considering that the concept of hypertext is quite different from the traditional paper-based paradigm with which students are already familiar, the finding of positive reactions to HF by the students suggests further use of hyperlinks for the presentation of on-line information. The students found the hypertext-based texts and reference aids helpful to their study of Korean in the reading component. They clearly preferred HF to NHF and noted that on-line information on vocabulary and grammar in a hypertext format is more beneficial than that in a non-hypertext format. This supports the suggestion that the linking characteristics of hypertext can assist FL learning by providing easy and immediate access to references.

The findings of this study draw out some theoretical and practical implications and point the way toward further research. First, it is important to look at the ways in which the content is presented as well as the content itself when designing or selecting CALL programs for reading. In the long term, HF might be more productive than PF, especially with the increasing need for reading on the computer screen. In the future, it may be more important to know how to optimize reading performance when text is displayed electronically. Also, hypertext seems to have considerable potential for enhancing self-management of learning and contributing to individual reading development. Second, there seems to be value in using hypertext applications that present on-line reference aids and provide reading texts and additional information on the same screens. This might suggest to CALL software developers that it would be advantageous to design more hypertext-based programs because students working in an individualized learning mode find hyperlinks useful; one might suggest the development of more hyper-linked reference tools such as learner dictionaries and culture dictionaries. Finally, this study responds to the demand for classroom-

REFERENCES

Armstrong, K.M. & Yetter-Vassot, C. (1994). Transforming teaching through technology. Foreign Language Annals, 27, 475-486.

Ashworth, D. (1996). Hypermedia and CALL. In M.C. Pennington (Ed.) The power of CALL (pp. 79-95). Houston, T.X.: Athelstan.

Aust, R., Kelley, M.J. & Roby, W. (1993). The use of hyper-reference and conventional dictionaries. Educational Technology Research and Development, 41 (4), 63-73.

Biber, D. (1992). Applied linguistics and computer applications. In W. Grabe & R.B. Kaplan (Eds), Introduction to applied linguistics (pp. 257-278). Reading, M.A.: Addison-Wesley.

Buzo, A. & Shin, G.-H. (1994). Learning Korean: New directions - Advanced study modules 1, 2 & 3. Melbourne: National Korean Studies Centre.

Chun, D.M. & Plass, J.L. (1996a). Effects of multimedia annotations on vocabulary acquisition. Modern Language Journal, 80, 183-198.

Chun, D.M. & Plass, J.L. (1996b). Facilitating reading comprehension with multimedia. System, 24, 503-519.

De Ridder, I. (2000). Are we conditioned to follow links? Highlights in CALL materials and their impact on the reading process. Computer Assisted Language Learning, 13 (2), 183-195.

Donaldson, R.P., Haggstrom, M.A. & Morgan, L.Z. (1994). Reading is creative, too! A discussion of two HyperCard stacks which can facilitate second-language reading comprehension. Computer Assisted Language Learning, 7, 195-208.

Edmonds, S. (1997, March). Making sense of hypertext. Paper presented at CAL 97, University of Exeter, United Kingdom.

Evans, M. (1993). Nicolas: Using HyperCard with intermediate level French learners. System, 21, 213-229.

Fox, J. (1990). A microcomputer-based approach to training in second language reading skills. Computer Assisted Language Learning, 1, 29-40.

Hulstijn, J.H. (1993). When do foreign-language readers look up the meaning of unfamiliar words? The influence of task and learner variables. Modern Language Journal, 77, 139-147.

Hult, S., Kalaja, M., Lassila, O. & Lehtisalo, T. (1990). HyperReader – An interactive course in reading comprehension. System, 18, 189-198.

Knight, S. (1994). Dictionary use while reading: The effects on comprehension and vocabulary acquisition for students of different verbal abilities. Modern Language Journal, 78, 285-299.

Landow, G.P. (1990). Popular fallacies about hypertext. In D.H. Jonassen & H. Mandl (Eds.) Designing hypermedia for learning (pp. 39-59). Berlin: Springer-Verlag.

Laufer, B. & Hill, M. (2000). What lexical information do L2 learners select in a CALL dictionary and how does it affect word retention? Language Learning & Technology, 3 (2), 58-76. Retrieved August 8, 2001, from http://llt.msu.edu/vol3num2/laufer-hill/index.html

Leffa, V.J. (1992). Making foreign language texts comprehensible for beginners: An experiment with an electronic glossary. System, 20, 63-73.

Lomicka, L.L. (1998). "To gloss or not to gloss": An investigation of reading comprehension online. Language Learning & Technology, 1 (2), 41-50. Retrieved August 8, 2001, from http://llt.msu.edu/vol1num2/article2/default.html

Lyman-Hager, M.A., Davis, J.N., Burnett, J. & Chennault, R. (1993). Une vie de boy: Interactive reading in French. In F.L. Borchardt & E.M.T. Johnson (Eds.) Proceedings of the CALICO 1993 annual symposium on "Assessment" (pp. 93-97). Durham, N.C.: Duke University.

McVicker, J. (1992). Several approaches to computer-based reading study. CAELL Journal, 3 (4), 2-11.

Sengupta, S. (1996). Creating a hypertext database to help Hong Kong English language teachers teach writing. System, 24, 187-198.

Slatin, J.M. (1991). Composing hypertext: A discussion of writing teachers. In E. Berk & J. Devlin (Eds.) Hypertext / Hypermedia handbook (pp. 55-64). New York: Intertext.

Son, J.-B. (1998). Understanding hypertext: A discussion for TEFL. English Teaching, 53 (3), 113-124.

Tolhurst, D. (1995). Hypertext, hypermedia, multimedia defined? Educational Technology, 35 (2), 21-26.

Whalley, P. (1993). An alternative rhetoric for hypertext. In C. McKnight, A. Dillon & J. Richardson (Eds.) Hypertext: A psychological perspective (pp. 7-17). London: Ellis Horwood.